Written by

Chong Hip Kuen, in Form 3

Acknowledgement
Thanks for IMO trainers Dr. Leung and Mr. T. W. Leung.
*This text is not going to be commercial.

First addition in 2011.


Caution: This passage is for fun only. Anyone who reads it may result of
addiction of Mathematics. Just take it not so serious for Mathematics.

Chapter 0 Introduction
What is combinatorics? Simply speaking, combinatorics is a study of
methods to do a thing. In problems of combinatorics, we need to count the
number of method for doing a thing correctly. It is not hard if you have good
method to do a problem.
What is probability? Probability is a number to indicate the likelihood or
chance for an event to happen. In problems of probability, sometimes we
need to find the numbers of favourable outcomes and possible outcomes. So
I think combinatorics and probability are two subjects of Mathematics which is
closely related.
In following chapters, let us play in the garden of Mathematics and find some
fruits from it.

Chapter 1 Simple Probability
So let us start with a simple question.

Question 1.1:
In a dark box there are 1 red ball, 4 green balls, and 5 blue balls. Assuming
the box is fair, find,
a) The possible outcome of the result of drawing balls from the box.
b) The probability of drawing a red ball from the box. (Just draw a ball from
the box)
c) The probability of drawing a green ball from the box.

Solution 1.1:
a) Obviously, the possible outcome of the result of drawing balls from the box
is 1+4+5=10.
b) The favourable outcome is 1. So the probability = 1/10.
c) The favourable outcome is 4. So the probability = 4/10 = 2/5.

It shows that probability of a thing = favourable outcome / possible outcome,
where all probability of outcomes should be the same.
In this chapter we will just discuss simple situation, i.e. [that is] we will not
discuss: find the probability of drawing a red ball or a green ball, etc.

A concept should be clear: for example, if we have calculated a probability, e.g.
1/2. It does not mean that if I have done it once, it will not have a chance to
appear again in the second time I do the same thing. That means we cannot
guess the next outcome by probability. Only when the outcome has shown,
the probability of the outcome becomes 1 and other outcomes become 0.
[If we use Chinese to explain it again, it will be:

]

Another thing we should learn in probability is expected values.
There are two types of expected values. They are expected number of
occurrences and expected value of a variable. May I use following
example to explain them for you.

Example 1.2:
If we toss a coin for 20 times, find the expected number of occurrences of
head.

Solution 1.2:
Since the probability of the coin is head is 1/2. We can say that expected
number of occurrences is (1/2)*20=10. This means we expect it occurs 10
times in 20 times.

Example 1.3:
If there are 4 $1 coin, 3 $2 coins, 2 $5 coins and 5 $10 coins in a dark bag, find
the expected value of a coin drawn from the bag.

Solution 1.3:
The expected value = 1*(4/14)+2*(3/14)+5*(2/14)+10*(5/14) = $5. This
means we expect the average value drawn from the bag in infinitely many time
from this bag is $5 if we put the coin into the bag after drawing it.

So that is the basic probability. Is that simple?

Chapter 1 Homework:
Question 1:
A number is chosen from the integers 1 to 100000000. Find the probability
that,
a) The chosen number is a multiple of 2
b) The chosen number is a multiple of 256
c) The chosen number is a multiple of 7.

Question 2:
There are 100 balls in a dark box. In them, 28 of them is red, 37 of them is
green, and 19 of them is blue. Find,
a) The probability of the colour of the chosen ball is red
b) The probability of the colour of the chosen ball is blue
c) The probability of the colour of the chosen ball is neither of above colour.

Question 3:
A fair dice is thrown once. Find the probability of getting a number
a) Which is between 1 and 6 (exclusive)
b) Which is a prime number
c) Which is a composite number
d) Which is not a prime number or a composite number
e) Which is a factor of 120
f) Which is a multiple of 8.

Question 4:
In a telephone number 25974543 (my telephone number ^^), find the
probability of getting a digit from this number
a) Which is a prime number
b) Which is a factor of 365
c) Which is a multiple of 1
d) Which its multiplicative inverse [x 1/x] is smaller than 0.5
e) Which has odd number of factors
f) Which when it is squared, the last digit of it is 2
g) Which when it is cubed, the first digit of it is 1.
Chapter 2 Simple Algorithm
After a chapter of introduce probability, would you become more familiar of
probability? So in this chapter, I am going to talk about conjunction of cases.
In probability, we cannot find numeric algorithm more complicated than
addition and multiplication.

Numeric algorithm (addition, multiplication)

Addition of probability: If we want to do things from criteria A or criteria B,
which have no case which satisfies criteria A and B, we can add up the
probability of criteria A and probability of criteria B. Let me use an example to
tell you about that.

Example 2.1:
If there is a dark box, which has 5 red balls, 3 green balls and 2 blue balls
inside. Find the probability of drawing a ball that is red or green.

Solution 2.1:
Why do you bold or and and in this passage? It is because these words
will cause confusion. Above and means it should satisfies both criteria A
and B. And the or means it can satisfies either A or B or both.
The favourable outcome = 5+3 = 8. (Since there is no ball which is both red
and green) The probability = 8/10 = 4/5.

Multiplication of probability: If we want to do thing which need two or more
consecutive steps, say P and Q. The numbers of methods to do P and Q is
p and q respectively. So the total number of methods = pq.

Example 2.2:
Tom wants to go to City C from City A. We know that Tom can go to City C by
cars or planes. There are 8 airlines he can choose from if one wants to travel
from City A to City C. From City A, there are only roads though City B to City
C. There are 5 roads from City A to City B, and 7 roads from City B to City C.
Find out the number of methods for Tom to travel from City A to City C.
Solution 2.2:
Obviously, total number of methods = # [number] of methods from air + # of
methods from land.
# of methods from air = 8 (given)
# of methods from land = 5*7 = 35 (Since he needs consecutive steps to go to
City C)

Can you find something I have missed? Yes, it is addition algorithm.
Recall: do things from criteria A or criteria B, no case which satisfies criteria A
and B, use addition.
So what should we do if there are some cases which satisfy some of the
criteria?
Let me introduce the first principle in this notes to you: Inclusion-Exclusion
Principle [].
Just let me give you an example to explain what that principle is.

Example 2.3:
In integers from 1 to 10, find the number of integers which
a) Can be divided by 2
b) Can be divided by 5
c) Can be divided by 10
d) Can be divided by 2 or 5

Solution 2.3:
a) 5, no explanation.
b) 2
c) 1
d) This part is the main point I want to talk about. If you are a careless
person, probably you will write this as a solution:
Since the question want us find integers which can be divided by 2 or
5, so we can add up the number of integers which can be divided by 2
and 5. Therefore, the answer is 2+5=7. Am I a clever boy?
No, you are not. Sorry about that. Why? It is because 10 can satisfy
both the case of integers which can be divided by 2 and 5. If you just add
them up, you will recount the 10. So, we should minus the recounting part.
The answer is 5+2-1=6.

This is what the principle means. If there is recount part, you can just minus it.
If there is part that we have not counted, you can add it onto the result.

For example, in 1 to 30, find number of integers which can be divided by 2 or 3
or 5. If we just do [# of 2] + [# of 3] + [# of 5] [# of 2 and 3 = 6] [# of 2 and 5
= 10] [# of 3 and 5 = 15], it will result occurrence of another clever boy.
Why?
Let me list out all the numbers. I will mark it red if it is # of 2, 3, 5; green if it is
# of 6, 10, 15; blue if it is # of 6 and 10 and 15 = 30.

2, 3, 4, 5, 6, 8, 9, 10, 12, 14, 15, 16, 18, 20, 21, 22, 24, 25, 26, 27, 28, 30.

So in above calculation, we have done:
{2,4,6,8,10,12,14,16,18,20,22,24,26,28,30} + {3,6,9,12,15,18,21,24,27,30} +
{5,10,15,20,25,30} {6,12,18,24,30} {10,20,30} {15,30}
= {2,3,4,5,6,8,9,10,12,14,15,16,18,20,21,22,24,25,26,27,28} and miss out 30.
It is because we have not counted it.

Therefore, we can have:
Sum of numbers satisfy one of criteria Sum of numbers satisfy two of criteria
+ Sum of numbers satisfy three of criteria Sum of numbers satisfy four of
criteria + - = Total number satisfy any of criteria.
This is Inclusion-Exclusion Principle.

That is all for algorithm in probability and combinatorics. So, we come to
logical algorithm.

Logical algorithm (and, or, not)
And and or are mentioned before. So I just skip it in this section.

Not algorithm
Some question is very difficult to calculate it directly, but we can invert it and
calculate it quickly.

Example 2.4:
Tom is playing Aeroplane chess (by Wikipedia). He rolls a dice six times and
gets at least one 6. Find the probability for Tom doing that.
Concept trap: In chapter one, we have talked about in guessing the coin. We
know that we cannot guess the next answer by previous answer. So in this
question also have the same case. A person rolls a dice for six times does
not indicate he can get a 6. It is not true.

Solution 2.4:
Okay, some of Secondary 3 students probably using a chart to solve this
question. This is not so wise because he should draw 46656 cells in order to
list out all the outcomes. However, it is also very hard to do it straight. So
what should we do?
Remember the not algorithm? Can we not the question?
So the question will be changed to: What is the probability for Tom to roll a dice
six times and cannot get any 6.
If you remember what the previous section is talking about, you know that this
question can use multiplication to finish.
Probability of Tom to roll a dice once and cannot get a 6 = (6-1)[not]/6 = 5/6.
Probability of Tom doing same thing 6 times = (5/6)^6 = 33.5%, 3 sig fig.
Probability of Tom not doing this kind of thing (Tom rolls a dice six times and
gets at least one 6) = not (33.5%) = 66.5%, 3 sig fig.

That is where not can be used in.

In this chapter, all the and, or, not, add, multiplication are not formal
wordings. So please use your own word to explain it and do not use this kind
of words.

Homework 2
1. Find numbers of integers smaller or equal to 500, and divisible by 2, 3, 5, or
7. [I.E.P. can be also used in combinatorics]
2. Find # of ways to separate 9 into sum of several numbers [including 9=9].
3. Find # of 10-digit number which all digits are not divisible by 2, 3 or 5.
4. In a 3x3 western-chessboard-like grids, find # of ways to move a piece of
chess from top-right hand corner to bottom-left hand corner. [Which it can
only move downwards and leftwards. ] [Consider how many methods is
there to get to each grid, and do some operation to get the answer. ]
5. In 314159265358, find numbers of digit that is divisible by 2 or divisible by 3
and 5.
6. There are 40 students in a class. Class teacher wants to choose Class
Representative, Class monitor and Chairman of the Class. However,
since both Class monitor and Chairman of the Class have lots of power,
class teacher prevents one to do both jobs. Find # of ways that class
teacher chooses students to do jobs.
7. Earth is going to be crushed by an asteroid which is 10000 cube kilometres!
There are only two ways to escape from the Earth. One is taking
spaceships which can support it going to outer space. In Earth, we only
have 15 of them. Another way to escape is to take spaceships which can
only support it going to the space station, and take another kind of
spaceships to outer space. The number of them is 20 and 15 respectively.
Find # of ways we can escape from the Earth.
8. If we roll a dice for 10 times, it has 1/1024 chance each output is even
numbers. Find the probability of rolling a dice for 10 times and cannot get
all outputs are odd numbers without using a calculator.
9. Find the probability for a person to win paper-scissor-rock for 3
consecutive times or lose them all. [Assume that person has only played
it for 3 times and there is tie in this game. ]
10. In integers 1-1000, find the probability of drawing a number which
a) Is divisible by 4 or 25
b) Is not divisible by both 4 and 25
c) Is divisible by 4 or 5 but not 10
d) Has a reminder not equal to zero when it is divided by 10
e) Has at least 1 zero in the number
11. By how many times rolling a dice will result more than 99% that at least 1
six will be resulted.
Chapter 3 Factorial and Permutation
This chapter probably is closer to the main part of combinatorics and
probability that the last three chapter. This is because after this chapter, you
will have knowledge to crack some simple MO [Mathematics Olympiad]
question. The world of combinatorics will also become somehow more
colourful than last three chapters just doing: find the probability of . And we
will not do: try to list all outcomes of such criteria or try to explain this result to
others. In this chapter, we will see: Find the number of methods to arrange
in criteria.

Just let me introduce a simple notation will be seen below and in the world.
The notation I want to introduce is factorial [].

n! = n factorial / factorial of n = n(n-1)(n-2)*3*2*1. In this formula, n have to
be an integer and 0 factorial = 0! = 1.
This notation will be used frequently in combinatorics and probably probability.
Its approximate value n! =
2

or ((2n)*n^n / e^n), where e=2.717(find

in calculator (

) or (e^x).

What is the use of it?

# of ways to arrange n different objects in a row
Just let me use an example to explain it to you.

Example 3.1:
If there are 3 boys we want to arrange their order, namely Alan, Ben, and Carry.
What is # of ways to arrange all of them?

Solution 3.1:
If we know nothing in combinatorics, we can still do this question straight:
Let Alan be A, Ben be B and Carry be C.
The methods to arrange all of them are listed below:
ABC, ACB, BAC, BCA, CAB and CBA.
Therefore, the # of ways to arrange all of them is 6.
This method is called exhaustion [] in combinatorics and it will be
discussed later after we have enough knowledge to do harder combinatorics.
However, this method has some possible problems.
First, if a careless person like me counts the method of doing things, the
answer probably will be +1 or -2. This leads to a wrong answer and it is
intolerable.
Second, if the number of classmates is not 3, probably 40. So do you want to
count it by exhaustion? The answer is about 8.16x10^47. It needs huge
amount of time and patience to finish the answer. (If you count it from the
start of the universe, you are still counting now! ) Is the Earth large? And
is an atom small? From estimation, Earth has about 1.33x10^50 atoms.
The answer is near # of atoms in the Earth! So you can see the number is too
large to use exhaustion.
Then what should we do? Just consider the three spaces for Alan, Ben and
Carry.
__ __ __ they should stand on the spaces on the left of this sentence.
In the first space, there are 3 choices of persons to stand on. Since the first
space uses 1 person, there are only 2 choices of persons to stand on. The
third one has only 1 choice, which is the left person. So the # of ways to
arrange them = 3*2*1 (consecutive ways to arrange them) = 3! = 6.
Similarly, if there are 40 persons for us to arrange, # of ways = 40! .
So, in here, we can summarise that:
There are n! ways to arrange n different objects in a row.

# of ways to arrange n different objects in a row of r spaces which objects
cannot be repeatedly placed
Just let me use another example to explain it to you.

Example 3.2:
If there are 5 girls we want to arrange their order, namely Ava, Betty, Cherry,
Dorothy, and Ellie. What is # of ways to choose Class Representative, Class
Monitor and Chairman of the Class which any two of them should not be done
by a person?

Answer 3.2:
By using exhaustion, we can list:
ABC, ABD, ABE, ACB, ACD, ACE, ADB, ADC, ADE, AEB, AEC, AED, etc.
We can count the number of ways to choose posts is 60.
Same problem happens here when there is 40 persons in the class.
So we will use the similar method as above.
__ __ __, there is 3 posts for them to compete.
The first post has 5 choices of people. The second post has 4, and the third
has 3. # of ways = 5*4*3 = 60.
However, how we can generalise the cases in this type of question?
We can see when we choose r posts from n persons, there is
n(n-1)(n-2)(n-r+1) ways to make there is r parts multiplied together.
We can make use of factorial to simplify it as n! / r!, i.e.
!
!
. We call this
formula as

=
!
!
(not in mainland China).
So, in here, we can summarise that:
There are

=
!
!
ways to arrange n different objects in a row of r spaces
which objects cannot be repeatedly placed.
Since this formula is so difficult to type, so in this passage, I will use nPr = n! / r!
to replace it.

Remember: We define that nP0 = 1.

# of ways to arrange n different objects in a row of r spaces which objects can
be repeatedly placed
If we change the Example 3.2 into:

Example 3.3:
If there are 5 girls we want to arrange their order, namely Ava, Betty, Cherry,
Dorothy, and Ellie. What is # of ways to choose Class Representative, Class
Monitor and Chairman of the Class which a person can do more than 1 post?
Solution 3.3:
If you have interest in exhaustion, you can try it by yourself. It is too long for
me to write on here.
Using the similar method, __ __ __, there is 3 posts for them to compete.
The first post has 5 choices of people. Since a person can do more than 1
post, the second post has 5 choices, and the third has 5 too. # of ways =
5*5*5 = 125.
We can generalise the formula as:
There are n^r ways to arrange n different objects in a row of r spaces
which objects can be repeatedly placed.
If you use them wisely, you can do many problems in combinatorics and
probability. Understanding knowledge thoroughly is only way to learn Maths!
Homework 3:
1. How many 3-digit positive integers have
a All three digits of same parity [, odd or even]
b All three digits different
c At least two digits equal
2. (PCIMC 2003) How many 4-digit positive integers have the product of their
digits even?
3. In a 6x6 square grid,
a How many squares can be found?
b How many rectangles can be found?
4. If we want to choose 5 persons out of a class of 40, and arrange their order.
How many methods are there to arrange their order if 5 persons are given?
5. If there is 5 persons, what is the difference of # of ways to arrange their
order and # of ways to pick them to do 5 posts which a person can do more
than 1 post?
6. In word aluminium, what is # of ways to
a Rearrange it with only its characters and all its characters.
b Rearrange it with criteria of (a) and not changing its vowels and
consonants.
c Rearrange it with criteria of (a) and not changing the sequences of
vowels and consonants.
d Rearrange it with criteria of (a) and min should be the head of the
word.
e Rearrange it with criteria of (a) and all the double letters should
appear in the beginning of the word.
f Rearrange it by throwing some letters away and form a palindrome.
g Rearrange it by throwing some letters away and form a word with no
repeating letters.
7. There are 40 students in a class. Find the number of ways to choose 3
identical posts from the students which a person at most can only do a
post.
8. If there is 5 boys and 3 girls in total in a class, and they are randomly
permutated. Find the probability of
a Leader (First one) is a boy and last one is a girl.
b Every girls front and back (if any) are boys.
c There are no girls between any 2 boys.
d There are no boys between any 2 girls.
e Permutation satisfies (c) and (d).
Chapter 4 Combination and Extension of Combination
After introducing function nPr, I think you will have a better concept about
combinatorics and probability. So, in this chapter I will continue to discuss
more function about combinatorics which can help you in probability. And in
next chapter I will discuss about formula of these functions.

# of ways to choose r objects from n objects and the order of them being
chosen is not important

Example 4.1:
There are 40 students in a class. Find the probability of choosing no. 2, 5, 37
as class monitors which are not distinct.

Solution 4.1:
The favourable outcome here obviously = 1.
However, what is # of possible outcomes?
Recall: nPr is the # of ways to choose r objects from n objects.
In this time, order is not important. So what is # of possible outcome?
Let say if the teacher has chosen 1, 2, 3 as monitors, so how can the teacher
choose them (order)?
They can be: 1,2,3 ; 1,3,2 ; 2,1,3 ; 2,3,1 ; 3,1,2 ; 3,2,1 .
There are 6 possible cases for choosing 1, 2, 3 as monitors.
We can observe that there are (r!) possible cases for choosing r things if we
count the order when counting cases.
So # of possible outcomes = 40P3 / 3! = 40! / (3! * 37!);
40!
(3!)(37!)
= 9880.
The probability of choosing no. 2, 5, 37 as class monitors is 1/9880.

We can see in general case, when we choose r objects from n objects and the
order of them being chosen is not important, the number of ways to choose
them = nPr / r! ;

!
= n!/(r!*(n-r)!) ;
!
!()!
. We have a symbol for it nCr ;

. nCr = n!/(r!*(n-r)!) ;.

=
!
!()!
.

There is another example of nCr.
Example 4.2:
Find probability of finding r 1s in an n-digit binary string.

Solution 4.2:
This is a simple question. Since we should choose r 1s in n 0 or 1, so the
favourable outcome = nCr. The possible outcome = 2^n.
Therefore, the probability = nCr / 2^n ;

.

# of ways to put r identical balls into n different boxes which may contain any #
of balls

Example 4.3:
There are 4 different kinds of dishes in a restaurant. There will be a discount
if you choose any 6 dishes among those 4 kinds. So what is # of ways to
choose dishes to have this discount?

Solution 4.3:
This is a more difficult question than above one, isnt it? There are so many
ways that we cannot list them out in a short time. So if you are a lazybones,
what should you do?
Just consider this:
The manager of this restaurant likes to write program very much. He has
written a program to identify what the kind of those 6 dishes:
A dish is represented as a 0 in this program and two kinds of dishes
are separated by a 1. The example output would be like this:





This is output string if you observe carefully is containing 6 0s
because there are 6 dishes in this restaurant. And there are 3 1s
because there are 3 separations from AB, BC, and CD.
So the combination of dishes = 9C3 = 84 (since in 9 digits, we choose 3 digits
to become 1).
Dish A Dish B Dish C Dish D String
3 2 1 0 000100101
0 0 6 0 110000001
5 0 0 1 000001110
By observation, in general case, # of ways to put r identical balls into n
different boxes which may contain any # of balls = (n+r-1)C(r).
In this time, we use a new symbol to indicate this kind of numbers. It is

; nHr. And the formula of this is

+1
=
(+1)!
(1)!!
; nHr = (n+r-1)C(r)
= (n+r-1)! / ((n-1)!*r!).

Example 4.4:
There are 4 distinct taxis and 8 distinct buses arranging in a row. Find the
probability of randomly selected row of vehicles has no two consecutive taxis.

Solution 4.4:
Let B symbolise non-distinct Bus, and T symbolise non-distinct Taxi.
To confirm that all Ts are separated from each other, the favourable cases will
all have *TB*TB*TB*T*, where * is a wildcard with any # of buses (include 0).
This structure makes all T are separated from at least a bus.
Let the number of bus in the first * to fifth * be a, b, c, d and e.
We know that a + b + c + d + e = 5, where a, b, c, d, e are non-negative
integers.
This is equivalent to put 5 identical 1s into a, b, c, d, and e five boxes.
Therefore, # of method if buses and taxis are respectively not distinguishable =
5H5 = 9C5 = 126.
Since all taxis and buses are distinguishable, so the number of favourable
outcomes = 126 * 4! * 8! = 121927680.
The number of possible outcome = 12! = 479001600.
The probability = 0.255, correct to 3 significant figures. Is it pretty large?

Exhaustion
Sometimes the question is harder and we cannot be lazy to do them. We
need to list all the possibility of the criteria systematically to count # of ways of
doing such thing. This kind of method is called exhaustion.

Example 4.5:
By 1, 3, 5, 7, 9, how many methods of using 2 of them to combine a two-digit
prime number (i.e. 1 and 3 13 is a prime number)?
Solution 4.5:
By exhaustion, i.e.13, 17, 19, 31, 37, 53, 59, 71, 73, 79, 91, 97 are prime
numbers. So the number of prime numbers = 12.
Homework 4:
1. In a restaurant, a set dinner consists of 5 dishes chosen from a collection of
7. How many different choices are there if
a) The 5 chosen dishes must not be the same?
b) The same dish can be chosen more than once?
2. What is the probability if we choose 13 cards randomly from a pack of 54 so
that at least two suit is missing?
3. (IMO HK Prelim 1999) A solid pyramid VABCD, with a quadrilateral base
ABCD, is to be coloured on each of the five faces such that no two faces
with a common edge will have the same colour. If five different colours
are available, what is the number of ways to colour the pyramid?
4. There is a 7x7 square grid. How many ways can one travel from the
bottom left-hand corner to top right-hand corner, if each step one can only
go upwards or rightwards?
5. In the Mark Six lottery in Hong Kong, what is the probability of winning
a) The first prize (6 balls)
b) The second prize (5.5 balls)
c) The third prize (5 balls)
d) The fourth prize (4.5 balls)
e) The fifth prize (4 balls)
f) The sixth prize (3.5 balls)
g) The seventh prize (3 balls)
h) By 2011/11, if you win the first prize, you will receive $29149350.
The second prize will give you $762020. Third - $88330. Fourth -
$9600. Fifth - $640. Sixth - $320. Seventh - $20. What is the
expected value of a $10 ticket?
6. There are four types of marbles, A, B, C, and D. There are 15 marbles of
each colour.
a) If we need to choose two marbles of different types, how many
choices are there?
b) If we need to choose 5 marbles, how many different combinations of
colours are there?
c) What will be the answer to (b) if we require at least 1 red marble?
7. What is the formula of
a) Order matters and repetition allowed?
b) Order matters and repetition not allowed?
c) Order does not matter and repetition allowed?
d) Order does not matter and repetition not allowed?